Computing the nearest diagonally dominant matrix

DOI10.1002/(SICI)1099-1506(199811/12)5:6%3C461::AID-NLA141%3E3.0.CO;2-VMaRDI QIDQ2713568zbMATH OpenFDO

Authors María Mendoza, M. Raydan, Pablo Tarazaga

Publication date 10 June 2001

Published in Numerical Linear Algebra with Applications (Search for Journal in Brave)

Full work available at URL https://doi.org/10.1002/(SICI)1099-1506(199811/12)5:6%3C461::AID-NLA141%3E3.0.CO;2-V

zbMATH Keywords

alternating projection algorithmdiagonally dominant matricesDykstra's algorithmKuhn-Tucker conditionsmatrix conesnumerical experiments

Mathematics Subject Classification ID

Computational methods for sparse matrices (65F50) Numerical mathematical programming methods (65K05) Quadratic programming (90C20)

Review text

The problem of minimizing the distance of a given matrix to the set of symmetric and diagonally dominant matrices with positive diagonal is considered. Using the Kuhn-Tucker condition the authors first characterize the projection onto the cone of diagonally dominant matrices with positive diagonal and then they apply \textit{R. L. Dykstra}'s alternating projection algorithm [J. Am. Stat. Assoc. 78, 837-842 (1983; Zbl 0535.62063)] to force symmetry and, thus, solve the problem. Two different implementations of Dykstra's algorithm applied to the given problem are suggested and and their behaviour on some numerical experiments is discussed.

Recommendations

• Selective alternating projections to find the nearest SDD\(^{+}\) matrix
• scientific article; zbMATH DE number 2076014
• Primal and polar approach for computing the symmetric diagonally dominant projection
• Computing a nearest symmetric positive semidefinite matrix
• Dykstra's Algorithm for a Constrained Least-squares Matrix Problem

Cites work

• A cyclic projection algorithm via duality
• A lower bound for the smallest singular value of a matrix
• A Modified Cholesky Algorithm Based on a Symmetric Indefinite Factorization
• A New Modified Cholesky Factorization
• A Nonlinear Programming Problem in Statistics (Educational Testing)
• A numerical procedure for finding the positive definite matrix closest to a patterned matrix
• An Alternating Projection Algorithm for Computing the Nearest Euclidean Distance Matrix
• Approximation by matrices positive semidefinite on a subspace
• Computing a nearest symmetric positive semidefinite matrix
• Cones of diagonally dominant matrices
• Dykstra's Algorithm for a Constrained Least-squares Matrix Problem
• Dykstra's algorithm for constrained least-squares rectangular matrix problems
• On the cone of positive semidefinite matrices
• Positive definite constrained least-squares estimation of matrices
• Positive Semidefinite Matrices: Characterization via Conical Hulls and Least-Squares Solution of a Matrix Equation
• Semi-Definite Matrix Constraints in Optimization
• The rate of convergence of dykstra's cyclic projections algorithm: The polyhedral case
• The structure of cones of matrices
• Theory of cones

Cited in

(6)

• Selective alternating projections to find the nearest SDD\(^{+}\) matrix
• Computational acceleration of projection algorithms for the linear best approximation problem
• Dykstra's algorithm with strategies for projecting onto certain polyhedral cones
• Diagonally Dominant Principal Component Analysis
• Primal and polar approach for computing the symmetric diagonally dominant projection
• An acceleration scheme for Dykstra's algorithm

This page was built for publication: Computing the nearest diagonally dominant matrix

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2713568)